# hardhat

## Introduction

hardhat is a *developer focused* package designed to ease the creation
of new modeling packages, while simultaneously promoting good R modeling
package standards as laid out by the set of opinionated [Conventions for
R Modeling
Packages](https://tidymodels.github.io/model-implementation-principles/).

hardhat has four main goals:

- Easily, consistently, and robustly preprocess data at fit time and
  prediction time with
  [`mold()`](https://hardhat.tidymodels.org/reference/mold.md) and
  [`forge()`](https://hardhat.tidymodels.org/reference/forge.md).

- Provide one source of truth for common input validation functions,
  such as checking if new data at prediction time contains the same
  required columns used at fit time.

- Provide extra utility functions for additional common tasks, such as
  adding intercept columns, standardizing
  [`predict()`](https://rdrr.io/r/stats/predict.html) output, and
  extracting valuable class and factor level information from the
  predictors.

- Reimagine the base R preprocessing infrastructure of
  [`stats::model.matrix()`](https://rdrr.io/r/stats/model.matrix.html)
  and [`stats::model.frame()`](https://rdrr.io/r/stats/model.frame.html)
  using the stricter approaches found in
  [`model_matrix()`](https://hardhat.tidymodels.org/reference/model_matrix.md)
  and
  [`model_frame()`](https://hardhat.tidymodels.org/reference/model_frame.md).

The idea is to reduce the burden of creating a good modeling interface
as much as possible, and instead let the package developer focus on
writing the core implementation of their new model. This benefits not
only the developer, but also the user of the modeling package, as the
standardization allows users to build a set of “expectations” around
what any modeling function should return, and how they should interact
with it.

## Installation

You can install the released version of hardhat from
[CRAN](https://CRAN.R-project.org) with:

``` r
install.packages("hardhat")
```

And the development version from [GitHub](https://github.com/) with:

``` r
# install.packages("pak")
pak::pak("tidymodels/hardhat")
```

## Learning more

To learn about how to use hardhat, check out the vignettes:

- [`vignette("mold", "hardhat")`](https://hardhat.tidymodels.org/articles/mold.md):
  Learn how to preprocess data at fit time with
  [`mold()`](https://hardhat.tidymodels.org/reference/mold.md).

- [`vignette("forge", "hardhat")`](https://hardhat.tidymodels.org/articles/forge.md):
  Learn how to preprocess new data at prediction time with
  [`forge()`](https://hardhat.tidymodels.org/reference/forge.md).

- [`vignette("package", "hardhat")`](https://hardhat.tidymodels.org/articles/package.md):
  Learn how to use
  [`mold()`](https://hardhat.tidymodels.org/reference/mold.md) and
  [`forge()`](https://hardhat.tidymodels.org/reference/forge.md) to help
  in creating a new modeling package.

You can also watch [Max Kuhn](https://github.com/topepo) discuss how to
use hardhat to build a new modeling package from scratch at the XI
Jornadas de Usuarios de R conference
[here](https://canal.uned.es/video/5dd25b9f5578f275e407dd88).

[![](https://i.imgur.com/XKIZfWd.png)](https://canal.uned.es/video/5dd25b9f5578f275e407dd88)

## Contributing

This project is released with a [Contributor Code of
Conduct](https://contributor-covenant.org/version/2/0/CODE_OF_CONDUCT.html).
By contributing to this project, you agree to abide by its terms.

- For questions and discussions about tidymodels packages, modeling, and
  machine learning, please [post on RStudio
  Community](https://forum.posit.co/new-topic?category_id=15&tags=tidymodels,question).

- If you think you have encountered a bug, please [submit an
  issue](https://github.com/tidymodels/hardhat/issues).

- Either way, learn how to create and share a
  [reprex](https://reprex.tidyverse.org/articles/articles/learn-reprex.html)
  (a minimal, reproducible example), to clearly communicate about your
  code.

- Check out further details on [contributing guidelines for tidymodels
  packages](https://www.tidymodels.org/contribute/) and [how to get
  help](https://www.tidymodels.org/help/).

# Package index

## Preprocessing

- [`mold()`](https://hardhat.tidymodels.org/reference/mold.md) : Mold
  data for modeling
- [`forge()`](https://hardhat.tidymodels.org/reference/forge.md) : Forge
  prediction-ready data

## Prediction

- [`spruce_numeric_multiple()`](https://hardhat.tidymodels.org/reference/spruce-multiple.md)
  [`spruce_class_multiple()`](https://hardhat.tidymodels.org/reference/spruce-multiple.md)
  [`spruce_prob_multiple()`](https://hardhat.tidymodels.org/reference/spruce-multiple.md)
  : Spruce up multi-outcome predictions
- [`spruce_numeric()`](https://hardhat.tidymodels.org/reference/spruce.md)
  [`spruce_class()`](https://hardhat.tidymodels.org/reference/spruce.md)
  [`spruce_prob()`](https://hardhat.tidymodels.org/reference/spruce.md)
  : Spruce up predictions
- [`quantile_pred()`](https://hardhat.tidymodels.org/reference/quantile_pred.md)
  [`is_quantile_pred()`](https://hardhat.tidymodels.org/reference/quantile_pred.md)
  [`extract_quantile_levels()`](https://hardhat.tidymodels.org/reference/quantile_pred.md)
  [`as_tibble(`*`<quantile_pred>`*`)`](https://hardhat.tidymodels.org/reference/quantile_pred.md)
  [`as.matrix(`*`<quantile_pred>`*`)`](https://hardhat.tidymodels.org/reference/quantile_pred.md)
  : Create a vector containing sets of quantiles

## Utility

- [`model_frame()`](https://hardhat.tidymodels.org/reference/model_frame.md)
  : Construct a model frame

- [`model_matrix()`](https://hardhat.tidymodels.org/reference/model_matrix.md)
  : Construct a design matrix

- [`model_offset()`](https://hardhat.tidymodels.org/reference/model_offset.md)
  : Extract a model offset

- [`delete_response()`](https://hardhat.tidymodels.org/reference/delete_response.md)
  : Delete the response from a terms object

- [`standardize()`](https://hardhat.tidymodels.org/reference/standardize.md)
  : Standardize the outcome

- [`new_model()`](https://hardhat.tidymodels.org/reference/new_model.md)
  : Constructor for a base model

- [`add_intercept_column()`](https://hardhat.tidymodels.org/reference/add_intercept_column.md)
  :

  Add an intercept column to `data`

- [`weighted_table()`](https://hardhat.tidymodels.org/reference/weighted_table.md)
  : Weighted table

- [`fct_encode_one_hot()`](https://hardhat.tidymodels.org/reference/fct_encode_one_hot.md)
  : Encode a factor as a one-hot indicator matrix

- [`impute_quantiles()`](https://hardhat.tidymodels.org/reference/impute_quantiles.md)
  :

  Impute additional quantiles from a `quantile_pred`

## Validation

- [`scream()`](https://hardhat.tidymodels.org/reference/scream.md) :
  Scream

- [`shrink()`](https://hardhat.tidymodels.org/reference/shrink.md) :
  Subset only required columns

- [`validate_column_names()`](https://hardhat.tidymodels.org/reference/validate_column_names.md)
  [`check_column_names()`](https://hardhat.tidymodels.org/reference/validate_column_names.md)
  :

  Ensure that `data` contains required column names

- [`validate_no_formula_duplication()`](https://hardhat.tidymodels.org/reference/validate_no_formula_duplication.md)
  [`check_no_formula_duplication()`](https://hardhat.tidymodels.org/reference/validate_no_formula_duplication.md)
  :

  Ensure no duplicate terms appear in `formula`

- [`validate_outcomes_are_binary()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_binary.md)
  [`check_outcomes_are_binary()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_binary.md)
  : Ensure that the outcome has binary factors

- [`validate_outcomes_are_factors()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_factors.md)
  [`check_outcomes_are_factors()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_factors.md)
  : Ensure that the outcome has only factor columns

- [`validate_outcomes_are_numeric()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_numeric.md)
  [`check_outcomes_are_numeric()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_numeric.md)
  : Ensure outcomes are all numeric

- [`validate_outcomes_are_univariate()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_univariate.md)
  [`check_outcomes_are_univariate()`](https://hardhat.tidymodels.org/reference/validate_outcomes_are_univariate.md)
  : Ensure that the outcome is univariate

- [`validate_prediction_size()`](https://hardhat.tidymodels.org/reference/validate_prediction_size.md)
  [`check_prediction_size()`](https://hardhat.tidymodels.org/reference/validate_prediction_size.md)
  : Ensure that predictions have the correct number of rows

- [`validate_predictors_are_numeric()`](https://hardhat.tidymodels.org/reference/validate_predictors_are_numeric.md)
  [`check_predictors_are_numeric()`](https://hardhat.tidymodels.org/reference/validate_predictors_are_numeric.md)
  : Ensure predictors are all numeric

## Blueprint

- [`default_formula_blueprint()`](https://hardhat.tidymodels.org/reference/default_formula_blueprint.md)
  [`mold(`*`<formula>`*`)`](https://hardhat.tidymodels.org/reference/default_formula_blueprint.md)
  : Default formula blueprint

- [`default_recipe_blueprint()`](https://hardhat.tidymodels.org/reference/default_recipe_blueprint.md)
  [`mold(`*`<recipe>`*`)`](https://hardhat.tidymodels.org/reference/default_recipe_blueprint.md)
  : Default recipe blueprint

- [`default_xy_blueprint()`](https://hardhat.tidymodels.org/reference/default_xy_blueprint.md)
  [`mold(`*`<data.frame>`*`)`](https://hardhat.tidymodels.org/reference/default_xy_blueprint.md)
  [`mold(`*`<matrix>`*`)`](https://hardhat.tidymodels.org/reference/default_xy_blueprint.md)
  : Default XY blueprint

- [`is_blueprint()`](https://hardhat.tidymodels.org/reference/is_blueprint.md)
  :

  Is `x` a preprocessing blueprint?

- [`new_formula_blueprint()`](https://hardhat.tidymodels.org/reference/new-blueprint.md)
  [`new_recipe_blueprint()`](https://hardhat.tidymodels.org/reference/new-blueprint.md)
  [`new_xy_blueprint()`](https://hardhat.tidymodels.org/reference/new-blueprint.md)
  [`new_blueprint()`](https://hardhat.tidymodels.org/reference/new-blueprint.md)
  : Create a new preprocessing blueprint

- [`new_default_formula_blueprint()`](https://hardhat.tidymodels.org/reference/new-default-blueprint.md)
  [`new_default_recipe_blueprint()`](https://hardhat.tidymodels.org/reference/new-default-blueprint.md)
  [`new_default_xy_blueprint()`](https://hardhat.tidymodels.org/reference/new-default-blueprint.md)
  : Create a new default blueprint

- [`refresh_blueprint()`](https://hardhat.tidymodels.org/reference/refresh_blueprint.md)
  : Refresh a preprocessing blueprint

- [`run_forge()`](https://hardhat.tidymodels.org/reference/run-forge.md)
  :

  [`forge()`](https://hardhat.tidymodels.org/reference/forge.md)
  according to a blueprint

- [`run_mold()`](https://hardhat.tidymodels.org/reference/run-mold.md) :

  [`mold()`](https://hardhat.tidymodels.org/reference/mold.md) according
  to a blueprint

- [`update_blueprint()`](https://hardhat.tidymodels.org/reference/update_blueprint.md)
  : Update a preprocessing blueprint

## Case Weights

- [`new_case_weights()`](https://hardhat.tidymodels.org/reference/new_case_weights.md)
  **\[experimental\]** : Extend case weights

- [`is_case_weights()`](https://hardhat.tidymodels.org/reference/is_case_weights.md)
  **\[experimental\]** :

  Is `x` a case weights vector?

### Importance Weights

- [`importance_weights()`](https://hardhat.tidymodels.org/reference/importance_weights.md)
  **\[experimental\]** : Importance weights

- [`new_importance_weights()`](https://hardhat.tidymodels.org/reference/new_importance_weights.md)
  **\[experimental\]** : Construct an importance weights vector

- [`is_importance_weights()`](https://hardhat.tidymodels.org/reference/is_importance_weights.md)
  **\[experimental\]** :

  Is `x` an importance weights vector?

### Frequency Weights

- [`frequency_weights()`](https://hardhat.tidymodels.org/reference/frequency_weights.md)
  **\[experimental\]** : Frequency weights

- [`new_frequency_weights()`](https://hardhat.tidymodels.org/reference/new_frequency_weights.md)
  **\[experimental\]** : Construct a frequency weights vector

- [`is_frequency_weights()`](https://hardhat.tidymodels.org/reference/is_frequency_weights.md)
  **\[experimental\]** :

  Is `x` a frequency weights vector?

## Setup

- [`create_modeling_package()`](https://hardhat.tidymodels.org/reference/modeling-usethis.md)
  [`use_modeling_deps()`](https://hardhat.tidymodels.org/reference/modeling-usethis.md)
  [`use_modeling_files()`](https://hardhat.tidymodels.org/reference/modeling-usethis.md)
  : Create a modeling package

## Information

- [`get_data_classes()`](https://hardhat.tidymodels.org/reference/get_data_classes.md)
  : Extract data classes from a data frame or matrix
- [`get_levels()`](https://hardhat.tidymodels.org/reference/get_levels.md)
  [`get_outcome_levels()`](https://hardhat.tidymodels.org/reference/get_levels.md)
  : Extract factor levels from a data frame

## Development

- [`tune()`](https://hardhat.tidymodels.org/reference/tune.md) : Mark
  arguments for tuning
- [`extract_workflow()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_recipe()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_spec_parsnip()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_fit_parsnip()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_fit_engine()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_mold()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_preprocessor()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_postprocessor()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_tailor()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_parameter_dials()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_parameter_set_dials()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  [`extract_fit_time()`](https://hardhat.tidymodels.org/reference/hardhat-extract.md)
  : Generics for object extraction

## Data

- [`hardhat-example-data`](https://hardhat.tidymodels.org/reference/hardhat-example-data.md)
  [`example_train`](https://hardhat.tidymodels.org/reference/hardhat-example-data.md)
  [`example_test`](https://hardhat.tidymodels.org/reference/hardhat-example-data.md)
  : Example data for hardhat

# Articles

### All vignettes

- [Forging data for
  predictions](https://hardhat.tidymodels.org/articles/forge.md):
- [Molding data for
  modeling](https://hardhat.tidymodels.org/articles/mold.md):
- [Creating Modeling Packages With
  hardhat](https://hardhat.tidymodels.org/articles/package.md):